arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2607.02998cs.CVcs.AIcs.LG

CONFLUX:一种用于3D胸部CT合成的潜在扩散模型及强化学习后训练

CONFLUX: A Latent Diffusion Model for 3D Chest-CT Synthesis with RL Post-Training

Department of Biomedical Data Science, Stanford University School of Medicine(斯坦福大学医学院生物医学数据科学系);Department of Mathematical Modelling, Statistics & Bioinformatics, Ghent University(根特大学数学建模、统计与生物信息学系);Department of Electrical Engineering, Stanford University(斯坦福大学电气工程系)

Max Van Puyvelde, Halil Ibrahim Gulluk, Wim Van Criekinge, Olivier Gevaert

AI总结:

提出CONFLUX潜在扩散模型用于胸部CT合成,由3D变分自编码器和整流流变压器构成,基于放射学元数据生成,通过强化学习后训练增强对临床属性控制,在三平面弗雷歇距离上领先并发布模型与数据集。

摘要:

3D医学图像的可控生成模型可合成具有特定临床属性的体积,但需要高保真、原生3D且符合条件的样本。我们提出CONFLUX,一种用于胸部计算机断层扫描(CT)的潜在扩散模型:一个3D变分自编码器压缩每个体积,一个整流流变压器在潜在空间中生成。通过自适应层归一化,生成以结构化放射学元数据(18个异常发现、性别、年龄和重建内核)为条件。该模型在三平面弗雷歇距离上领先强大的体积基线(MAISI的FID为32.3对74.6),同时展示了对临床属性的直接控制。为了加强这种控制,我们添加了一个在线强化学习后训练阶段(组相对策略优化),该阶段奖励分类器从每个生成的体积中恢复所需发现的可靠程度。由单独的独立分类器判断,后训练消除了相对于真实扫描可靠性的47%的不足。我们发布了该模型和一个约20万个合成胸部CT数据集,其条件元数据涵盖了各种临床发现。

英文摘要:

Controllable generative models of 3D medical images can synthesize volumes with specified clinical attributes, but this demands samples that are simultaneously high-fidelity, natively 3D, and faithful to the requested conditioning. We present CONFLUX, a latent diffusion model for chest computed tomography (CT): a 3D variational autoencoder compresses each volume, and a rectified-flow transformer generates in the latent space. Generation is conditioned on structured radiological metadata (18 abnormality findings, sex, age, and reconstruction kernel) through adaptive layer normalization. The model leads strong volumetric baselines on tri-planar Frechet distance (FID 32.3 vs. 74.6 for MAISI) while exposing direct control over clinical attributes. To strengthen that control we add an online reinforcement-learning post-training stage (group-relative policy optimization) that rewards how reliably a classifier recovers the requested findings from each generated volume. Judged by a separate, independent classifier, post-training removes 47% of the shortfall relative to real-scan reliability. We release the model and a ~200k synthetic chest-CT dataset with conditioning metadata spanning a wide variety of clinical findings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑